----- Original Message -----
> From: "Ezequiel Garcia" <ezequiel@xxxxxxxxxxxxxxxxxxxx>
> Sent: Saturday, November 22, 2014 6:55:42 PM
>
> On 11/08/2014 04:11 PM, Aaron Sierra wrote:
> > From: Jordan Friendshuh <jfriendshuh@xxxxxxxxxxx>
> >
> > Support the generic nand-ecc-mode and nand-ecc-strength device-tree
> > properties with the Freescale UPM NAND driver.
> >
> > This patch preserves the default software ECC mode while adding the
> > ability to use BCH ECC for larger NAND devices.
> >
> > Signed-off-by: Jordan Friendshuh <jfriendshuh@xxxxxxxxxxx>
> > Signed-off-by: Aaron Sierra <asierra@xxxxxxxxxxx>
> > ---
> > v2:
> > * Now using ECC mode and strength helpers from of_mtd.h
> > * ECC mode and strength checking is more robust
> >
> > .../devicetree/bindings/mtd/fsl-upm-nand.txt | 2 +
> > drivers/mtd/nand/Kconfig | 1 +
> > drivers/mtd/nand/fsl_upm.c | 51
> > +++++++++++++++++++---
> > 3 files changed, 49 insertions(+), 5 deletions(-)
> >
> > diff --git a/Documentation/devicetree/bindings/mtd/fsl-upm-nand.txt
> > b/Documentation/devicetree/bindings/mtd/fsl-upm-nand.txt
> > index fce4894..a9906f6 100644
> > --- a/Documentation/devicetree/bindings/mtd/fsl-upm-nand.txt
> > +++ b/Documentation/devicetree/bindings/mtd/fsl-upm-nand.txt
> > @@ -18,6 +18,8 @@ Optional properties:
> > - chip-delay : chip dependent delay for transferring data from array to
> > read registers (tR). Required if property "gpios" is not used
> > (R/B# pins not connected).
> > +- nand-ecc-mode : as defined by nand.txt ("soft" and "soft_bch", only).
> > +- nand-ecc-strength : as defined by nand.txt.
> >
>
> The nand.txt recommends that each binding documents the way these
> properties are interpreted.
Ezequiel,
I assume you are referring to this statement, "implementations are
encouraged to further specify the value(s) they support."
That is the only statement I found requesting documentation. I believe
that request is satisfied, since I listed all of the modes that the driver
would support with my patch applied. I used mxc-nand.txt as a reference.
> Moreover, there's no nand-ecc-step usage? Are you assuming a 512-byte size?
The ecc.bytes value is currently left to be calculated by the
NAND_ECC_SOFT_BCH case in nand_base.c. That calculation assumes a 512-byte
step size.
-Aaron
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
